Heuristic Analysis for Ecommerce: Get Quick Wins Before Quantitative Research

Heuristic Analysis

The conversion optimization process consists of multiple steps that overlap and interlock (heuristic analysis is just one of them). Many of these steps are based on data and research, so your initial step should always be to set up your data-gathering tools.

Heuristic analysis is Step Zero of any conversion optimization process worth its salt.

But where do you start?

Unless the website you want to optimize has obvious technical difficulties, the first step is to visually inspect the website. Do this by going through the website page by page, and try to spot obvious issues that could hinder conversion.

Just a few examples of potential conversion minefields:

  • Long forms with unnecessary fields: These add friction
  • Unintuitive menu navigation: Users may not know where to click
  • Unclear value propositions: Users don’t know what you’re offering
Heuristic analysis - long form
Here is an example of a very long form

This part of the process relies entirely on the researcher’s knowledge and experience. It is not based on any sort of data; instead, it relies on established best practices and what “usually” works on similar websites.

Important note: While heuristic analysis often precedes and informs other methods of research, it cannot and must not be your sole way of gathering data.

Why should you do heuristic analysis?

Considering the established fact that conversion optimization is a data-driven process, why is heuristic analysis a part of CRO at all?

It seems counterintuitive to talk about data-driven processes, gathering quantitative data and so on, but then begin with a method that is based only on experience. So why do we do it?

The answer is both simple and surprising. Heuristic analysis is the most cost-effective way to uncover a website’s low-hanging conversion optimization opportunities.

If a researcher can spend two or three hours simply observing and clicking around a website, and spot issues that can be fixed relatively easily and without using any quantitative tools or performing lengthy qualitative research… then by all means, heuristic analysis should precede other methods that take more time and resources.

What metrics is heuristic analysis based on?

In a nutshell, heuristic analysis focuses on usability.

The widely used analytical methodology developed by Jakob Nielsen, one of the founders of Nielsen Norman Group, consists of a list of usability heuristics. It’s often recognized as the best method for heuristic analysis of websites, software interfaces, and other types of user-to-machine interaction.

Here are the NNG usability indicators that apply to websites:

  • Ability to keep track of the process flow by user
  • Clarity and relevance
  • Freedom and control (the visitor’s ability to maintain control of the process and pursue their own objectives)
  • Consistency of site message
  • Prevention of errors by anticipating the most common errors, and reducing the possibility of triggering them by visitor action
  • Error messages tested for clarity and relevance in order to help users diagnose and recover from errors
  • Recognition
  • Flexibility and efficiency tests to allow visitors to use shortcuts and time-saving measures
  • Aesthetics and functional design
  • Documentation and help services

Let’s go through these one by one.

User’s ability to keep track of the process flow

This metric refers to the website user’s ability to instantly estimate their current “position” or location on the website relative to their desired objective or starting point. The visitor is made aware of their progress and the estimated number of remaining steps (or amount of time) before they reach the objective or goal.

A form that features some sort of progress indicator, especially for multiple-page forms, is a good example. The visitor is informed how many pages are left in the form, how many questions they have completed, how many questions remain, or simply shown the percentage of the form completed so far.

Heuristic analysis - easy to follow form
Here is an example of a form that informs prospects of their progress

In a broader sense, breadcrumb navigation and similar menu solutions serve the same purpose.

A website that fails in this respect will probably suffer from a high bounce rate, low conversion rate, and low visitor interaction. Visitors will feel lost without the ability to maintain control of their navigation through the site.

Clarity and relevance

Estimating and improving your website’s clarity and relevance is one of the primary tasks of conversion optimization.

Clarity refers to the visitor’s ability to understand a website’s message, while relevance refers to his ability to relate the product or service being offered to his own needs. These two concepts are critical indicators of a site’s success. It’s well known that if the visitor does not understand a website within five seconds, they will not proceed with their visit.

Clarity is the factor that helps visitors make the most out of this short amount of time. The website must clearly show and/or explain its purpose within those five seconds. So, in designing your website and writing its content, let clarity be your guiding principle.

Relevance, on the other hand, represents the connection between the visitor’s need and the content of the website. If a visitor perceives the site content as personally relevant, his motivation will increase, and he will pay more attention.

The way the human mind perceives information is also helpful to consider. There are many ways to convey information, but the best is imagery — since the brain processes pictures up to 60,000 times faster than words.

So the more visual the content your visitor sees on first glance, the more likely it is to grab his attention. Odds are that images or graphics will help make your content clearer.

Heuristic analysis - visual elements relevance
A visual representation of relevance. Image source

Freedom and control

People like to feel like they’re in control, and this applies to websites, too.

If your visitor feels like they’re being forced down a certain path on your website or that they cannot control the way they navigate the site, you will lose them.

To be clear, this doesn’t mean that your website cannot have a structure or ideal user flow. It must have those — but in order to instill a feeling of control in visitors, the site structure must be constructed as a logical progression.

Websites take freedom and control from visitors in a few different ways, including:

  • Frequent pop-ups
  • Slider banners or carousel-type catalogues that users cannot control
  • A lack of clear navigation

An example would be a “Buy Now!” pop-up that covers the visitor’s screen before they even have time to examine or consider the website’s value proposition or offer. Heuristic analysis should uncover these types of issues and inspire ideas about how to rectify them.

In short, any type of content that interrupts the visitor’s navigation, forces certain actions on them, or pushes a call to action before they are ready to convert will make a user feel like they have no control.

Heuristic analysis - popups example
Heuristic analysis – popups example (image source)

Consistency of the message

When a visitor lands on the website, he should be “greeted” by a clear message that serves as the site’s basic value proposition. This message should be omnipresent throughout the website.

Take Apple, for example. A consistent message runs through every apple.com page dedicated to a specific product. iPad pages consistently point out the message that the iPad is better and faster than any laptop.

But the consistency doesn’t stop there. Apple maintains this consistency in all of its marketing channels. That way, the visitor is primed for conversion long before they even get to the website. If your website offers a clear and relevant message that is consistent throughout, it should enjoy high conversion rates (provided there are no other issues with the site).

Heuristic analysis - consistency
Heuristic analysis – consistency

Error management and prevention

You can be certain that in navigating your e-commerce website, your visitor will occasionally make an error — whether it’s due to a misunderstanding or simple oversight.

Most frequently, errors happen when visitors need to fill in a form or type a comment or something similar. A good user interface aims to eliminate the possibility of error. When this is not possible, it aims to minimize the effort a user must expend to understand and fix the error.

Error notifications should be constructed in a way that informs the visitor of their error but doesn’t blame or scare them. Error messages should be informative and contain the best way to correct the error.

For example, many website forms require a visitor to enter their date of birth. Since date formats differ all over the world, the required format should be clearly pointed out in the field question. However, should the visitor still enter the wrong format, the ensuing error message should inform them that the format is wrong, and once again point out the correct format.

(Of course, the best approach is to eliminate the possibility of this type of error by designing the input field in a way that compels or guides the visitor to fill it out correctly the first time.)

Heuristic analysis - error message
Here is an example of a successful implementation of error prevention.

An example of a bad error message would be “You have entered a date in the wrong format!” Not only does the error message language blame the visitor, it unduly alarms and frustrates them by using ambiguous language.

Error recovery

Further improving the error messaging involves testing error messages for clarity and relevance. Users should be able to understand the meaning of the error message and how to fix the error — so give them clear instructions in plain language.

Avoid error messages that just supply an error code, since these have no helpful meaning outside of your development staff.

Recognition

If your website is designed to serve a returning audience (and most ecommerce sites are), it should be able to recognize returning customers and ease their navigation accordingly.

Let’s look at Amazon. Once you register on the site, your actions are tracked and used to improve your experience. After you buy something, you will start seeing similar and related items on the homepage. If you maintain a wishlist, Amazon will point out the items from your wishlist that are currently on sale.

Google search is another example. If you are logged into your Google account, when you start typing a search keyword, Google will offer keywords similar to those you previously searched for.

This type of recognition both makes it easier for a customer to use your site, and makes it feel more familiar. And, as you may know, it’s worth the effort — since returning customers frequently have a much higher conversion rate than new visitors.

Personalizing your website goes a long way toward creating a loyal customer base. As you do your heuristic analysis, dedicate special attention to potential personalization options and current personalization implementation.

Without personalization, the only alternative left for the returning user is their own memory. Since memory can be unreliable, by using personalization, you avoid having to force your visitors to remember the specifics of your website and the easiest ways to access the content they want.

Flexibility and efficiency of use

This heuristic is related to the previous heuristic. Customers who frequently visit a website develop their own unique way of navigating it and consuming its content.

Since it is impossible to guess the number of ways in which a visitor might interact, or the sequence of their actions, a well-designed site will offer ways for visitors to take shortcuts to their most frequently used calls to action and/or content. This can speed up the user interaction and further increase the customer’s feelings of belonging and familiarity.

Aesthetics and functional design

In addition to being clear and relevant, a well-designed website should also be functional.

By that, we mean that the information it contains is presented in an easily accessible way that minimizes the effort needed to get information.

Heuristic analysis - aesthetics & functional design
Heuristic analysis – aesthetics & functional design

For example, copy should contain only information that is necessary and relevant. Every unnecessary addition to the content draws visitors’ attention away, and can harm your store’s conversion rate.

Documentation and help

As you know by now, a well-designed website should be intuitive to use. Most contemporary ecommerce sites use a familiar layout just for this reason. Visitors react better to a familiar interface, so this isn’t a good place to experiment, unless you’re extremely confident that your new design is better-suited to the nature of your business.

Should your website actually require some explanation for its use, provide these directions right where they’re necessary. If steps need to be completed in a certain order or input provided in a specific format, inform users when they initiate the action.

Also, provide an outlet for users to ask for help, whether it’s a live chat or another form of direct contact. (A bonus of allowing this kind of customer outreach is that it’s incredibly valuable for use in qualitative research, so no website should be without it.)

How to conduct your heuristic analysis

Due to its extreme subjectivity at certain points, heuristic analysis is best conducted as a team — since multiple sets of eyes can help eliminate subjective observations and biases.

Plus, the more people who conduct an analysis, the less likely it is that they’ll overlook issues.

In fact, just 5 people will most likely spot 75% of all heuristic issues on a website:

Heuristic analysis - Source: Nielsen Norman Group
Heuristic analysis – Source: Nielsen Norman Group

Go forth and analyze!

Heuristic analysis is the most cost-effective way to spot potential improvement options for your e-commerce store. And it often doesn’t require much testing to implement those changes — since a focus on improving usability and user experience can only result in boosting your website’s performance.

Since heuristic analysis doesn’t require the collection of quantitative data or the conduction of user surveys, it should be done prior to any other research activity. Of course, you’ll need to work with your client and/or outside research consultants, since some of the findings of your analysis may depend on the nature of the business activity and the target audience.

Heuristic analysis also has another benefit: it allows you to better understand your ecommerce customers, enhancing their ease of use and enjoyment of the shopping process.

Heuristics are usually most conspicuous in their absence. However, if you don’t notice problems immediately, look over some user session recordings. These can shed light on where your website is failing users in one regard or another.

Published by

Edin is a Senior CRO Consultant. Edin is into Google Analytics and testing (any A/B testing tool really) and likes to write about it. You can follow Edin on Twitter.